supplementary materials for "four billion people facing severe water scarcity" by mesfin m. mekonnen and arjen y. hoekstra. the pdf includes a supplementary discussion, figure s1 showing average monthly blue water scarcity at a spatial resolution of 30 × 30 arc min, and table s1 comparing results between the current study and previous studies. the supplementary discussion summarizes assumptions from previous studies that estimated water scarcity. previous studies measured water scarcity as natural runoff per capita, with 1000 m³/cap/y as a threshold for severe scarcity. however, this measure does not account for the fact that water footprints may not be in the same basin as where people live. most studies estimated water scarcity annually, while two studies (hanasaki et al. and hoekstra et al.) considered environmental flow requirements. the latter two studies measured water scarcity differently, with hanasaki et al. focusing on crop water requirements. previous studies underestimated water scarcity due to large spatial units, annual rather than monthly basis, and exclusion of environmental flow requirements. the current study uses high spatial and temporal resolution to better capture true water scarcity, considering environmental flow requirements and water consumption rather than gross abstraction. the threshold for severe water scarcity is set at 0.4, meaning annual withdrawal should not exceed 40% of annual runoff. water consumption is 60% of total withdrawal, so annual consumption should not exceed 24% of annual runoff. the current study defines water scarcity (ws) as the fraction of water available for consumptive activities that has been consumed. water withdrawal refers to gross abstraction, while water consumption refers to net abstraction, also called blue water footprint. actual runoff is natural runoff minus upstream water consumption.supplementary materials for "four billion people facing severe water scarcity" by mesfin m. mekonnen and arjen y. hoekstra. the pdf includes a supplementary discussion, figure s1 showing average monthly blue water scarcity at a spatial resolution of 30 × 30 arc min, and table s1 comparing results between the current study and previous studies. the supplementary discussion summarizes assumptions from previous studies that estimated water scarcity. previous studies measured water scarcity as natural runoff per capita, with 1000 m³/cap/y as a threshold for severe scarcity. however, this measure does not account for the fact that water footprints may not be in the same basin as where people live. most studies estimated water scarcity annually, while two studies (hanasaki et al. and hoekstra et al.) considered environmental flow requirements. the latter two studies measured water scarcity differently, with hanasaki et al. focusing on crop water requirements. previous studies underestimated water scarcity due to large spatial units, annual rather than monthly basis, and exclusion of environmental flow requirements. the current study uses high spatial and temporal resolution to better capture true water scarcity, considering environmental flow requirements and water consumption rather than gross abstraction. the threshold for severe water scarcity is set at 0.4, meaning annual withdrawal should not exceed 40% of annual runoff. water consumption is 60% of total withdrawal, so annual consumption should not exceed 24% of annual runoff. the current study defines water scarcity (ws) as the fraction of water available for consumptive activities that has been consumed. water withdrawal refers to gross abstraction, while water consumption refers to net abstraction, also called blue water footprint. actual runoff is natural runoff minus upstream water consumption.